Understanding volunteering intentions with computational text analysis

• In Event: Quantitative perspectives on Volunteering

Abstract

Survey research on volunteerism generally focused on the characteristics of the active volunteer force in the specific time period of the study itself. This provides little guidance as to potential future engagement; here, we explore how computational techniques can be used to induce volunteering intentions from responses to open-ended survey questions.
Existing survey questions on this are limited. Individuals may be asked to rank volunteering out of a list of pre-given options. For instance panel members in the UK’s Understanding Society longitudinal surveys ranked voluntary work as the least important among “a list of things that some people say are important about retirement”.
Such survey questions say little about the way people frame and understand future volunteer intentions. Hustinx et al. (2022) criticize the dominant (survey-based) tradition of research on volunteering for treating volunteering as a homogeneous and stable entity. Such considerations support more inductive research methods which starts from how volunteering is defined, framed and contextualized by people themselves. In this spirit, we map the volunteering intentions of citizens within their specific context, unhindered by prefixed conceptualizations and embracing the benefits of both quantitative and qualitative research methods.
The UK’s National Child Development Study (NCDS) tracks a cohort of 12000 individuals born in 1958; it regularly asks questions about social participation. In the 2008 survey wave, when respondents were 50, they were asked to write a short free-text response to a question about how they envisioned their life at age 60; 7378 completed this task. This generated a considerable corpus of textual material.
In this study we explore different computational techniques to map volunteer intentions based on these open answers. While this project is work in progress, preliminary analyses already showed the usefulness of different approaches. Topic modelling seemed the most suitable technique for our purposes but did not yield meaningful topics. A second option is supervised machine learning, which turned out to perform adequately in terms of precision and accuracy. Manual inspection of the output gives additional insights in what the model did and did not pick up.
Besides providing novel insights on the potential of the post-war generations (‘Boomers’) as a possibly untapped pool of volunteers, this study makes a broader contribution to the methodological knowledge in our field. We are not aware of any study that used computational text analysis to distinguish volunteering intentions (either when using the open-ended answers of the NCDS (Elliott (2022) and Weber (2021)) or when classifying types of volunteers(Santelli et al., 2020).
By describing our methodological journey and reflecting on the pros and cons of different choices, we help the field to better understand the potential of computational text analysis in volunteerism research. We explicitly discuss the things that did not work, too, in order to also show the limitations of this method. Furthermore, by adopting a more open and inductive approach towards what ‘volunteering’ can mean for individuals, our study contributes to the academic discussion on the conceptualization and operationalization of volunteering.

References

Elliott, J. (2022). The Craft of Using NVivo12 to Analyze Open-Ended Questions: An Approach to Mixed Methods Analysis. The Qualitative Report, 27(6), 1673–1687. https://doi.org/10.46743/2160-3715/2022.5460

Hustinx, L., Grubb, A., Rameder, P., & Shachar, I. Y. (2022). Inequality in Volunteering: Building a New Research Front. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 33(1), 1–17. https://doi.org/10.1007/s11266-022-00455-w

Rooney, P. M., Steinberg, K., & Schervish, P. G. (2004). Methodology Is Destiny: The Effect of Survey Prompts on Reported Levels of Giving and Volunteering. Nonprofit And Voluntary Sector Quarterly, 33(4), 628–654.

Santelli, F., Ragozini, G., & Musella, M. (2020). What Volunteers Do? A Textual Analysis of Voluntary Activities in the Italian Context. In D. F. Iezzi, D. Mayaffre, & M. Misuraca (Eds.), Text Analytics (pp. 265–276). Springer International Publishing. https://doi.org/10.1007/978-3-030-52680-1_21

Weber, M. (2021). How Do 50-Year-Olds Imagine Their Future: Social Class and Gender Disparities. SAGE Open, 11(4), 215824402110615. https://doi.org/10.1177/21582440211061567

Wiepking, P. (2021). The Global Study of Philanthropic Behavior. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 32(2), 194–203. https://doi.org/10.1007/s11266-020-00279-6

Authors

• Elisabet Doodeman, Vrije Universiteit Amsterdam

• Arjen de De Wit, Vrije Universiteit Amsterdam

• John Mohan, University of Birmingham